Sensitivity Mapping Simulation Visualization and Info

[Jammy2211]

The Simulate object used in sensitivity mapping now receives a simulate_path which is a folder where simulation specific visualization and info can be output.

In the example below, I use this to output the dataset visuals, tracer visuals and a .json containing the tracer used to simulate the data:

class SimulateImaging:
def init(self, mask, psf):

    self.mask = mask
    self.psf = psf

def __call__(self, instance, simulate_path):
    """
    Set up the `Tracer` which is used to simulate the strong lens imaging, which may include the subhalo in
    addition to the lens and source galaxy.
    """
    tracer = al.Tracer.from_galaxies(
        galaxies=[
            instance.galaxies.lens,
            instance.perturbation,
            instance.galaxies.source,
        ]
    )

    """
    Set up the grid, PSF and simulator settings used to simulate imaging of the strong lens. These should be tuned to
    match the S/N and noise properties of the observed data you are performing sensitivity mapping on.
    """
    grid = al.Grid2DIterate.uniform(
        shape_native=self.mask.shape_native,
        pixel_scales=self.mask.pixel_scales,
        fractional_accuracy=0.9999,
        sub_steps=[2, 4, 8, 16, 24],
    )

    simulator = al.SimulatorImaging(
        exposure_time=300.0,
        psf=self.psf,
        background_sky_level=0.1,
        add_poisson_noise=True,
    )

    dataset = simulator.via_tracer_from(tracer=tracer, grid=grid)
    dataset.apply_mask(mask=self.mask)

    self.output_info(simulate_path=simulate_path, dataset=dataset, tracer=tracer)

    """
    The data generated by the simulate function is that which is fitted, so we should apply the mask for 
    the analysis here before we return the simulated data.
    """
    return dataset

def output_info(self, simulate_path, dataset, tracer):

    mat_plot = aplt.MatPlot2D(
        output = aplt.Output(path=simulate_path, format="png")
    )

    dataset_plotter = aplt.ImagingPlotter(
        imaging=dataset, mat_plot_2d=mat_plot
    )
    dataset_plotter.subplot_dataset()

    tracer_plotter = aplt.TracerPlotter(
        tracer=tracer, grid=dataset.grid, mat_plot_2d=mat_plot
    )
    tracer_plotter.subplot_lensed_images()

    with open(os.path.join(simulate_path, "tracer.json"), "w+") as f:
        json.dump(tracer.dict(), f, indent=4)

[codecov]

Codecov Report

Merging #711 (bcf9348) into main (560ecca) will increase coverage by 0.02%.
The diff coverage is n/a.

@@            Coverage Diff             @@
##             main     #711      +/-   ##
==========================================
+ Coverage   82.00%   82.02%   +0.02%     
==========================================
  Files         180      180              
  Lines       13264    13265       +1     
==========================================
+ Hits        10877    10881       +4     
+ Misses       2387     2384       -3     

see 2 files with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[rhayes777]

I feel like this breaks separation of concerns

[Jammy2211]

I feel like this breaks separation of concerns

I agree, but the simulation function is user-written and the quantities which are visualized depend on it, so I dont see an obvious way to avoid it.

[rhayes777]

I feel like this breaks separation of concerns

I agree, but the simulation function is user-written and the quantities which are visualized depend on it, so I dont see an obvious way to avoid it.

Fair enough